JP2014108728A - Vehicle body sideslip angle estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle body sideslip angle estimation device capable of improving accuracy in estimating a vehicle body sideslip angle at the time of steering on a low-friction coefficient road or in a tire nonlinear region.SOLUTION: In calculating a vehicle body sideslip angle on the basis of a lateral velocity estimation value and a longitudinal velocity estimation value obtained by detecting lateral acceleration, longitudinal acceleration, yaw rate and longitudinal velocity, and by inputting a lateral acceleration sensor value, a longitudinal acceleration sensor value and a yaw rate sensor value in to an observer 11 based on a lateral motion equation of the vehicle and a longitudinal motion equation thereof, the lateral acceleration sensor value and the longitudinal acceleration sensor value are respectively corrected on the basis of an error between the longitudinal velocity sensor value and the longitudinal velocity estimation value.

Description

  The present invention relates to a vehicle body side slip angle estimating device.

  Patent Document 1 discloses a technique for estimating a vehicle body side slip angle by combining a linear observer estimation method and a direct integration method.

JP-A-8-332934

The estimation method using the linear observer has a problem that the estimation error is large on a low μ road and a tire non-linear region where the vehicle slip angle estimation is highly demanded. On the other hand, the direct integration method operates accurately regardless of changes in road surface μ and tire characteristics, but there is a problem that the estimated value drifts because the influence of sensor noise and offset is accumulated by integral calculation.
An object of the present invention is to provide a vehicle body side slip angle estimating device capable of improving the estimation accuracy of a vehicle body side slip angle at the time of steering in a low μ road or a tire non-linear region.

  In the present invention, the lateral speed estimation value and the longitudinal speed obtained by inputting the lateral acceleration detection value, the longitudinal acceleration detection value, and the yaw rate detection value to the observer based on the lateral motion equation and the longitudinal motion equation of the vehicle. When obtaining the vehicle body side slip angle from the estimated value, the lateral acceleration detection value and the longitudinal acceleration detection value are corrected based on the error between the longitudinal speed detection value and the longitudinal speed estimation value.

  Therefore, since it is not affected by the modeling error, it is possible to reduce the estimation error on the low μ road and the tire nonlinear region. In addition, since the longitudinal acceleration detection value and the lateral acceleration detection value are mutually corrected by the lateral speed estimation value and the longitudinal speed estimation value, drift of the estimation value can be suppressed. Furthermore, since the lateral acceleration detection value and the longitudinal acceleration detection value are corrected by feedback of the error between the longitudinal speed detection value and the longitudinal speed estimation value, drift of the estimation value can be suppressed. As a result, it is possible to improve the estimation accuracy of the vehicle body side slip angle at the time of steering on a low μ road or a tire nonlinear region.

1 is a diagram illustrating a schematic configuration of a vehicle according to a first embodiment. It is a figure which defines a rigid body motion and coordinates when a vehicle motion is regarded as a rigid body motion in a three-dimensional space. FIG. 3 is a block diagram of an observer used for vehicle body side slip angle estimation according to the first embodiment. 3 is a flowchart showing a flow of vehicle side skid angle estimation processing of the vehicle running state estimation device 1. It is a time chart of the side slip angle when a lane change is performed on a low μ road. 6 is a yaw rate sensor value correction map according to a second embodiment. It is a time chart of the body side slip angle at the time of going straight. 12 is a flowchart illustrating a flow of yaw rate sensor value correction processing according to the third embodiment. It is a figure which shows the yaw rate sensor value correction | amendment effect | action of Example 3. FIG. 6 is a time chart of a vehicle body side slip angle according to a third embodiment.

[Example 1]
[overall structure]
FIG. 1 is a diagram illustrating a schematic configuration of a vehicle according to a first embodiment. The vehicle of the first embodiment is an FR drive type vehicle that drives rear wheels 10RL, 10RR with braking / driving motors 9RL, 9RR and steers with front wheels 10FL, 10FR.
The vehicle includes a vehicle running state estimation device 1, a yaw rate sensor (yaw rate detection means) 2, a lateral acceleration sensor (lateral acceleration detection means) 3, a longitudinal acceleration sensor (longitudinal acceleration detection means) 4, a wheel speed sensor (longitudinal speed) Detection means) 5, braking / driving motor ECU (Electronic Control Unit) 8, and braking / driving motors 9RL, 9RR.
The vehicle travel state estimation device 1 estimates the travel state of the vehicle such as the vehicle body side slip angle based on the detection results of the sensors. The yaw rate sensor 2 detects the yaw rate of the vehicle. The lateral acceleration sensor 3 detects the lateral acceleration of the vehicle. The longitudinal acceleration sensor 4 detects the longitudinal acceleration of the vehicle. The wheel speed sensor 5 detects the wheel speed of each wheel. The braking / driving motor ECU 8 controls the braking / driving motors 9RL and 9RR based on the operation input to the accelerator pedal 6 and the brake pedal 7 and the traveling state estimated by the vehicle traveling state estimating device 1.

ただし、変数上のドットは微分値を表す。各記号は以下の通りである。
α_x:前後方向加速度
α_y:横方向加速度
u:前後方向速度
v:横方向速度
r:ヨーレイト（上下軸周り角速度）
m:剛体の質量
F_x:前後方向外力
F_y:横方向外力 [Body slip angle estimation]
The vehicle side slip angle estimation in the vehicle running state estimation device 1 will be described.
Conventional vehicle side slip angle estimation needs to solve difficult problems such as modeling of nonlinear characteristics of tires, reliable estimation of road surface μ, removal of sensor noise and drift, etc. Realization was difficult. Therefore, in the vehicle body side slip angle estimation of the first embodiment, the vehicle motion is regarded as a rigid body motion in a three-dimensional space, and modeling as shown in FIG. 2 is performed. x is the front-rear direction, y is the horizontal direction, and z is the vertical direction.
Equation (1) shows the inertial acceleration in the longitudinal and lateral directions of the rigid body shown in FIG.

However, the dot on the variable represents the differential value. Each symbol is as follows.
α _x : longitudinal acceleration α _y : lateral acceleration
u: longitudinal speed
v: Lateral speed
r: Yaw rate (Angular velocity around vertical axis)
m: Mass of rigid body
F _x : External force in the longitudinal direction
F _y : Lateral external force

各記号は以下の通りである。
g:重力加速度
φ:x軸周りの回転角度
θ:y軸周りの回転角度
式(1),(2)をまとめると、

となり、各加速度G_x,G_yで状態u,vが駆動されると共に、重力に伴う外乱が加わった系とみなすことができる。 Incidentally, in equation (1), the vertical velocity and the angular velocity around the x and y axes are neglected. On the other hand, the magnitudes G _x and G _y of the sensor values of the lateral acceleration sensors 4 and 3 before and after mounting on the vehicle are expressed by Expression (2).

Each symbol is as follows.
g: Gravitational acceleration φ: Rotation angle around x axis θ: Rotation angle around y axis Equations (1) and (2) can be summarized as follows:

Thus, the states u and v are driven by the accelerations G _x and G _y , and the system can be regarded as a system to which disturbance due to gravity is added.

ただし、添え字tは時刻を表し、変数上のバーは状態方程式に基づく予測値であることを示す。各記号は以下の通りである。
G_t:予測に関するヤコビアン
H_t:計測更新に関するヤコビアン
K_t:カルマンゲイン
R_t:プロセスノイズの共分散
Q_t:観測ノイズの共分散
Σ_t:状態量の共分散
dt:演算更新時間
u_{t_sensor}:車速センサ値 Expression (3) is a time-varying nonlinear system including a yaw rate in the system matrix, and an extended Kalman filter is configured as follows for this system. The disturbances in the longitudinal and lateral directions due to gravity are expressed as w ₁ and w ₂ , respectively.

The subscript t represents time, and the bar on the variable indicates a predicted value based on the state equation. Each symbol is as follows.
G _t : Jacobian about prediction
H _t : Jacobian about measurement update
K _t : Kalman gain
R _t : Process noise covariance
Q _t : Observation noise covariance Σ _t : State quantity covariance
dt: Calculation update time
u _{t_sensor} : _Vehicle speed sensor value

A part of equation (9) is shown in a block diagram in FIG. FIG. 3 is a block diagram of an observer used for vehicle body side slip angle estimation according to the first embodiment. The observer 11 inputs the yaw rate sensor value, the lateral acceleration sensor value, the longitudinal acceleration sensor value, and the vehicle speed sensor value (the average value of the left and right rear wheel speed sensor values), and the longitudinal speed estimated value and the lateral speed estimated value. Is output.
The observer 11 includes a lateral speed estimator (lateral speed estimator) 12, a front-rear direction speed estimator (front-rear direction speed estimator) 13, and a front-rear direction speed estimation error compensator (acceleration detection value corrector) 14. The lateral speed estimation unit 12 calculates a longitudinal speed estimation value and corresponds to a conventional direct integration method. In the direct integration method, it can be seen that the sensor offset directly affects the estimated value. The front-rear direction speed estimation unit 13 calculates a lateral direction speed estimation value. As is apparent from FIG. 3, the lateral speed estimation value calculated by the lateral speed estimation unit 12 is obtained by multiplying the longitudinal speed estimation value calculated by the longitudinal speed estimation unit 13 by the yaw rate sensor value. For the calculation of the longitudinal speed estimation value by the longitudinal speed estimation section 13, a value obtained by multiplying the lateral speed estimation value calculated by the lateral speed estimation section 12 by the yaw rate sensor value is used. The longitudinal speed estimation error compensation unit 14 calculates the longitudinal acceleration sensor value and the lateral acceleration sensor value by a feedback amount obtained by multiplying the deviation (error) between the vehicle speed sensor value and the longitudinal speed estimation value by gains k ₁ and k _2. to correct. The gains k ₁ and k ₂ are components of the Kalman gain K _t . That is, optimum gains k ₁ and k ₂ are set by the extended Kalman filter.
The vehicle body side slip angle β is calculated with reference to the following equation (11) based on the estimated lateral speed value and the estimated longitudinal speed value estimated by the observer 11.

[Body slip angle estimation process]
FIG. 4 is a flowchart showing a flow of the vehicle body side slip angle estimation process of the vehicle running state estimation device 1, and each step will be described below.
In step S101, a longitudinal acceleration sensor value is input from the longitudinal acceleration sensor 4.
In step S102, a lateral acceleration sensor value is input from the lateral acceleration sensor 3.
In step S103, the yaw rate sensor value is input from the yaw rate sensor 2.
In step S104, the average value of the rear wheel speed sensor values is input from the wheel speed sensor 5 as the y vehicle speed sensor value.
In step S105, the front-rear direction speed estimated value and the lateral direction speed estimated value are calculated by the observer 11 shown in FIG.
In step S106, the vehicle body side slip angle is calculated from the longitudinal speed estimated value and the lateral speed estimated value using equation (11). Step S106 corresponds to vehicle side slip angle calculation means.

Next, the operation will be described.
[Improved accuracy of side slip angle estimation during steering]
In various vehicle motion control systems, it is necessary to estimate the vehicle side slip angle with high accuracy. Conventional representative methods are:
1. Estimation method using a linear observer using a vehicle model
2. Broadly divided into direct integration methods.
In the estimation method using the linear observer, a highly accurate estimated value can be obtained if the vehicle model is accurate. However, sufficient accuracy cannot be obtained when modeling errors such as road surface μ and vehicle weight are large. In general, the demand for vehicle side slip angle estimation is high on low-μ roads and tire non-linear regions, and therefore has weak points in highly demanded regions.
On the other hand, the direct integration method does not have a vehicle model and directly uses an in-vehicle sensor signal, so that in principle, it operates accurately regardless of changes in road surface μ and tire characteristics. However, since integral calculation is used, the effects of sensor noise and offset are accumulated, and the estimated value drifts.

On the other hand, in the vehicle running state estimation device 1 of the first embodiment, the observer 11 is configured using not only a lateral motion equation but also a longitudinal motion equation based on a direct integration method without a vehicle model.
Therefore, since it is not affected by the modeling error, it is possible to reduce the estimation error on the low μ road and the tire nonlinear region. In addition, since the longitudinal acceleration sensor value and the lateral acceleration sensor value are mutually corrected using the yaw rate sensor value, drift of the estimated value can be suppressed.
Further, since the lateral acceleration sensor value and the longitudinal acceleration sensor value are corrected by the feedback correction amount based on the error between the longitudinal speed sensor value and the longitudinal speed estimation value, the sensor noise and the offset are large. Also, the drift of the estimated value can be suppressed. At this time, in the first embodiment, since the extended Kalman filter for setting the gains k ₁ and k ₂ for determining the feedback correction amount is configured, each acceleration sensor value can be corrected in consideration of the influence of gravitational acceleration or the like. The vehicle side slip angle can be accurately estimated even while traveling.

  FIG. 5 is a time chart of the vehicle body side slip angle when a lane change is performed on a low μ road. In the estimation method using a linear observer, sufficient estimation accuracy is not obtained due to modeling errors. Further, the direct integration method cannot cope with continuous steering such as lane change because the estimated value drift increases due to accumulation of errors. On the other hand, in the first embodiment, even if continuous steering is performed, drift of the estimated value can be suppressed, and highly accurate vehicle slip angle estimation is realized.

Next, the effect will be described.
In Example 1, the following effects are exhibited.
(1) A lateral acceleration sensor 3 that detects the lateral acceleration of the vehicle, a longitudinal acceleration sensor 4 that detects the longitudinal acceleration of the vehicle, a yaw rate sensor 2 that detects the yaw rate of the vehicle, and a longitudinal velocity of the vehicle. Calculates the estimated lateral speed of the vehicle based on the lateral motion equation of the vehicle from the wheel speed sensor 5, the lateral acceleration sensor value, the yaw rate sensor value, and the estimated longitudinal speed value calculated by the longitudinal speed estimation unit 13. The vehicle's longitudinal speed estimation unit 12 estimates the vehicle's longitudinal speed based on the vehicle's longitudinal motion equation from the longitudinal acceleration sensor value, the yaw rate sensor value, and the lateral speed estimation value calculated by the lateral speed estimation unit 12. Vehicle body side slip angle calculating means for calculating a vehicle body side slip angle based on the lateral direction speed estimated value and the front and rear direction speed estimated value (S106) And a longitudinal speed estimation error compensator 14 that corrects the lateral acceleration sensor value and the longitudinal acceleration sensor value based on the error between the longitudinal speed sensor value and the longitudinal speed estimation value, respectively.
As a result, it is possible to improve the estimation accuracy of the vehicle body side slip angle at the time of steering on a low μ road or a tire nonlinear region.

(2) The longitudinal speed estimation error compensation unit 14 includes an extended Kalman filter that determines the correction amount of the lateral acceleration sensor value and the longitudinal acceleration sensor value based on the error between the longitudinal speed sensor and the longitudinal speed estimation value, respectively. Prepared.
As a result, the vehicle body side slip angle can be accurately estimated even when traveling on a slope or a cant road.

(3) Detects lateral acceleration, longitudinal acceleration, yaw rate, and longitudinal velocity, and uses the lateral acceleration sensor value, longitudinal acceleration sensor value, and yaw rate sensor value as observers based on the vehicle's lateral motion equation and longitudinal motion equation. When determining the vehicle slip angle from the estimated lateral speed value and the estimated longitudinal speed value obtained by inputting to 11, the lateral acceleration sensor value is calculated based on the error between the longitudinal speed sensor value and the estimated longitudinal speed value. The longitudinal acceleration sensor value is corrected.
As a result, it is possible to improve the estimation accuracy of the vehicle body side slip angle at the time of steering on a low μ road or a tire nonlinear region.

[Example 2]
The second embodiment is different from the first embodiment in that the yaw rate sensor value is corrected to a constant value other than zero when the yaw rate sensor value is near zero.
In the vehicle running state estimation device 1 according to the second embodiment, as shown in FIG. 6, when the yaw rate sensor output (yaw rate sensor value) is in a region near zero, the yaw rate sensor value is corrected to a constant value. FIG. 6 is a yaw rate sensor value correction map according to the second embodiment. The constant value is a positive value when the yaw rate sensor value is a positive value, and a negative value when the yaw rate sensor value is a negative value. Here, the area near zero of the yaw rate sensor value is a range of the yaw rate sensor value detected while the vehicle is traveling straight, and is a range wider than the yaw rate sensor noise width (for example, about 3% of the sensor full scale). ).
The vehicle running state estimation device 1 calculates the yaw rate used for the calculation from the yaw rate sensor value based on the map of FIG. 6, and inputs the calculated yaw rate to the observer 11 to estimate the vehicle body side slip angle.

Next, the operation will be described.
[Drift suppression during straight travel]
FIG. 7 is a time chart of the vehicle body side slip angle when traveling straight. When the lateral acceleration sensor value is offset, the direct integration method cannot cope with continuous steering such as lane change because the estimated value drift increases due to the accumulation of errors. On the other hand, in the method of the first embodiment, the accuracy of the side slip angle is improved during steering, but when the vehicle travels straight with the lateral acceleration sensor value being offset, the estimated value drifts as in the direct integration method. There is. This is because the yaw rate becomes zero when traveling straight, and thus the feedback effect by the Kalman gain is lost in the observer shown in FIG. 3, and the influence of the sensor offset cannot be suppressed.
In other words, since the key to connect the longitudinal and lateral motion equations of the vehicle is that the yaw rate sensor value is not zero, in the second embodiment, when the yaw rate sensor value is near zero, the yaw rate used for calculation is a constant value other than zero. Fixed to a value. Thereby, since the longitudinal and lateral motion equations can continue to be virtually connected, the estimation error drift is slightly increased in the region where the yaw rate is near zero, but the drift of the estimated value can be suppressed.

Next, the effect will be described.
In Example 2, in addition to the effects (1) to (3) of Example 1, the following effects are obtained.
(4) When the yaw rate sensor value is near zero, there is provided yaw rate detection value correction means (vehicle running state estimation device 1) for correcting the yaw rate sensor value to a constant value other than zero.
Thereby, the drift of the estimated value at the time of going straight can be suppressed.

Example 3
The third embodiment is different from the second embodiment in that hysteresis is set to a constant value for correcting the yaw rate sensor value.
FIG. 8 is a flowchart illustrating a flow of yaw rate sensor value correction processing according to the third embodiment.
In step S201, the front wheel front-rear direction μ gradient is acquired.
In step S202, it is determined whether or not the rear wheel front-rear μ gradient is near zero. If YES, the process proceeds to step S205, and if NO, the process proceeds to step S203. Here, the vicinity of zero is a value that can be taken by the longitudinal μ gradient of the rear wheel when the yaw rate sensor value is in a region near zero.
In step S203, the yaw rate sensor value is used as the yaw rate used in the calculation.
In step S204, the yaw rate used in the calculation is input to the observer 11.
In step S205, it is determined whether the yaw rate sensor value is changing in the increasing direction (from negative to positive). If YES, the process proceeds to step S206. If NO (positive to negative direction), step S207 is performed. Proceed to
In step S206, the yaw rate is set to a negative constant value A.
In step S207, the yaw rate is set to a positive constant value B.

Next, the operation will be described.
[Hunting prevention]
Since the yaw rate sensor value goes back and forth between zero points when going straight, the method of the second embodiment, which switches the value of the constant value between positive and negative, has a problem that the yaw rate used for calculation is hunting in the vicinity of straight ahead. Therefore, in the third embodiment, when the yaw rate sensor value changes in the increasing direction during straight traveling, the yaw rate used for the calculation is set to a negative constant value A, and when the yaw rate sensor value changes in the decreasing direction, the yaw rate used for the calculation is changed. A positive constant value B is assumed. As a result, as shown in FIG. 9, even when the yaw rate sensor value increases or decreases across the zero point when going straight, it is possible to prevent the yaw rate used for calculation from hunting.
FIG. 10 is a time chart of the vehicle body side slip angle according to the third embodiment, which can eliminate the influence of sensor drift regardless of whether the vehicle is steering or traveling straight, and realizes highly accurate vehicle body side slip angle estimation.

Next, the effect will be described.
In Example 3, in addition to the effects (1) to (3) of Example 1 and the effect (4) of Example 2, the following effects are produced.
(5) Hysteresis was set to a constant value.
Thereby, it is possible to prevent hunting of the yaw rate used for the calculation in the vicinity of the straight traveling, and it is possible to improve the estimation accuracy of the vehicle body side slip angle during the straight traveling.

(Other examples)
As mentioned above, although the form for implementing this invention was demonstrated based on the Example, the concrete structure of this invention is not limited to an Example, The design change of the range which does not deviate from the summary of invention And the like are included in the present invention.
For example, the yaw rate may be estimated from the vehicle speed and the steering angle.

1 Vehicle running state estimation device
2 Yaw rate sensor (Yaw rate detection means)
3 Lateral acceleration sensor (lateral acceleration detection means)
4 Longitudinal acceleration sensor (Measuring means for longitudinal acceleration)
5 Wheel speed sensor (front-rear direction speed detection means)
6 Accelerator pedal
7 Brake pedal
8 Braking drive ECU
9RL, 9RR Braking drive motor
10FL, 10FR Front wheel
10RL, 10RR Rear wheel
11 Observer
12 Lateral speed estimator (transverse speed estimator)
13 Front-rear direction speed estimation unit (front-rear direction speed estimation means)
14 Longitudinal speed estimation error compensator (acceleration detection value correction means)

Claims (5)

Lateral acceleration detection means for detecting lateral acceleration of the vehicle;
Longitudinal acceleration detection means for detecting longitudinal acceleration of the vehicle;
A yaw rate detecting means for detecting the yaw rate of the vehicle;
Longitudinal speed detection means for detecting the longitudinal speed of the vehicle;
Lateral speed estimation for calculating a lateral speed estimation value of the vehicle based on a lateral motion equation of the vehicle from the lateral acceleration detection value, the yaw rate detection value and the longitudinal speed estimation value calculated by the longitudinal speed estimation means Means,
A front-rear direction for calculating a vehicle front-rear direction speed estimation value based on a vehicle front-rear direction motion equation from the front-rear direction acceleration detection value, the yaw rate detection value, and the lateral direction speed estimation value calculated by the lateral direction speed estimation means Speed estimation means;
Vehicle body side slip angle calculating means for calculating a vehicle body side slip angle based on the lateral direction speed estimated value and the longitudinal speed estimated value;
Acceleration detection value correction means for correcting the lateral acceleration detection value and the longitudinal acceleration detection value based on an error between the longitudinal speed detection value and the longitudinal speed estimation value, respectively.
A vehicle body side slip angle estimating device comprising: In the vehicle body slip angle estimating device according to claim 1,
The vehicle body side slip angle estimation device according to claim 1, wherein the acceleration detection value correction means includes an extended Kalman filter that determines correction amounts of the lateral acceleration detection value and the longitudinal acceleration detection value based on the error. In the vehicle body slip angle estimating device according to claim 1 or 2,
A vehicle slip angle estimating device comprising a yaw rate detection value correcting means for correcting the yaw rate detection value to a constant value other than zero when the yaw rate detection value is near zero. The vehicle body slip angle estimating device according to claim 3,
A vehicle body side slip angle estimating device characterized in that hysteresis is set to the constant value.   Lateral acceleration, longitudinal acceleration, yaw rate, and longitudinal velocity are detected, and the lateral acceleration detection value, longitudinal acceleration detection value, and yaw rate detection value are input to the observer based on the vehicle lateral motion equation and the longitudinal motion equation. When the vehicle body side slip angle is obtained from the estimated lateral speed value and the estimated longitudinal speed value obtained in the above, the detected lateral acceleration value and the estimated longitudinal speed based on the error between the detected longitudinal speed value and the estimated longitudinal speed value. A vehicle body side slip angle estimation device that corrects each direction acceleration detection value.

Images